Tracking discrepancies between expected and actual flows of tourists in an urban destination: An application of user-generated data

In tourism studies, new means of data collection (e.g., smartphone and Internet) have opened up new opportunities for unveiling spatial patterns at various scales. This paper aims to analyze the discrepancy between actual and expected flows between scenic spots in an urban destination at different distance ranges, on an urban scale, based on the individual travel tips data extracted from online sources. The results based on Nanjing, China, indicate that scenic spots with a high degree of centrality are more important in the network of expected flow compared to actual flow. Actual and expected flows of scenic spots are the consequence of the interaction among various factors, including inflow degree, outflow flow, and distance. Moreover, these flows are distance-sensitive at a low level, with a higher sensitivity of the actual flow. Because of the actual situation during the travel, tourists may adjust the itinerary from the expected flow and avoid the trip between two distant scenic spots due to various constraints. Lastly, implications are provided for destination planning and marketing.